Grade how stable a system is from evidence

Grades each issue of a system from its sources, separating closed, prevented, repaired, rolled out and observed. For a team that must say honestly how stable something is.

What it asks for

  • about (Optional, words): What this run is for, in the person's words.
  • known (Optional, words): What is already known, decided or done, so the run starts from it instead of asking again.

The steps your AI walks

  1. Read how the person wants it written, and the method
  2. Build a ledger of sources, never chat memory
  3. Separate closing a task from being stable
  4. Score each issue and let critical issues gate the stage
  5. Generate the report from data and check it
  6. Try to disprove it, then record and hand over

Use it from your AI

Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this workflow by what it does and runs it.

Say to your AI: “Use Trillion to grades each issue of a system from its sources, separating closed, prevented, repaired, rolled out and observed. For a team that must say honestly how stable something is”